Throttle response concern-
First, let's look at the throttle response different between the pipe sizes. I'll take a WAG at some absolute numbers, just to make my point.
8ft of 2.5in piping is about 30L of volume
8ft of 3in piping is about 44L of volume
The turbo is probably flowing at least 120L/s of pressurized air (I did some fuzzy math for this number, 500hp requires around 750ft^3/min of atmospheric air, divided by 3 for the pressure boost, and converted to L/s)
basically that 14L difference would take less than 0.1s to fill. There's no reason to fuss about 100ms of throttle response, especially in a street car.
Piping loss concern-
Here is a good article that explains flow lesses in a somewhat simple manor:
http://engineering-references.sbainvent.com/fluid-mechanics/head-loss.php#.UYLkB7WR-So
... although the equations can look a little intimidating, we're just looking at their characteristics, not compute them exactly.
Major head loss - pressure drop due to friction with the pipe wall
Minor head loss - pressure drop due to fittings, turns, or transitions in the piping
There are a couple charts that show inlet and outlet minor head loss factors. Basically, if the pipe diameter transitions are abrupt or poorly designed they will induce much more minor head loss than the major head loss they reduce. Not that they can't be done properly, but it takes care and thought to do it right.
As for the major head loss from making the pipe smaller, let's do some basic math:
flow volume [m^3/s] = flow velocity [m/s] * flow area [m^2]
Let's assume a fixed amount of HP requires requires a fixed amount of air flow. So we keep flow volume fixed and look at flow area and flow velocity:
flow velocity3 * flow area3 = flow velocity2.5 * flow area2.5 = flow volume
algebra...
flow area3 / flow area2.5 = flow velocity2.5 / flow velocity3
pi*1.5^2/pi*1.25^2 = 1.44 = velocity2.5 / velocity3
That is, the velocity of the air in the 2.5in piping will be 1.44 times as fast a the velocity of air in the 3in piping.
Note equation (3) in the article where head loss is expressed as a function of fluid velocity squared.
with the velocity increasing by a factor of 1.44 the head loss will be (1.44)^2 = 2.1, thus the head loss through the piping will be over twice as much, both major and minor losses.
What does this head loss mean? It means a difference in pressure between the turbo outlet and the intake manifold. If the car was losing 2psi with 3in pipe, it will be losing 4psi with 2.5in pipe, but I don't know the absolute number.
What does this mean for performance? The turbo has to output the same flow rate at a higher pressure to create the same power. This pushes the operating point of the turbo straight up on a compressor chart, generally making it less efficient. It also heats the intake air more and restricts the exhaust more.
It would be unlikely that the engine will make the same power given all of these sad side-effects of restricting the intake with smaller piping.
Now, let's all be a little reasonable here, I'm not saying everything will melt and explode and kill a bunch of babies, just that everything will be a little hotter and a little more stressed and it will cost you power.
/scienced